Proliferative diabetic retinopathy is characterized by obstruction of retinal blood vessels and retinal ischemia leading to release of factors that stimulate neovascularization. One factor that plays a major role is vascular endothelial growth factor (VEGF). Transgenic mice with photoreceptor-specific expression of VEGF develop neovascularization originating from the deep capillaries of the retina, but not from the superficial capillaries or choroidal blood vessels. The deep capillary bed develops after the superficial capillary bed and choroidal vessels and is still developing at the time that the transgene expression begins; because of its developmental stage, the deep capillary bed may be more responsive to VEGF. We hypothesize that overexpression of VEGF alone is not sufficient to stimulate new vessel growth from mature blood vessels, but rather the vessels must be made more responsive to VEGF by other factors that are expressed during development or in ischemic retina. We propose to test this hypothesis using the tetracycline responsive inducible transactivation system in conjunction with retina-specific promoters in transgenic mice. We will determine if the timing or location of VEGF expression is a critical determinant of the pattern of neovascularization. We will also explore whether hypoxia-inducible factor-1 (HIF-1), Angiopoietin 1 (Ang1) or Ang2 play a role in the development of retinal neovascularization. These studies could provide new insights into mechanisms by which retinal neovascularization occurs and identify new targets for therapeutic intervention.